Deep Learning Applications in Physical Education: A Systematic Review of Educational and Sport Science Perspectives

DOI

10.2991/978-2-38476-591-1_30How to use a DOI?

Keywords

artificial intelligence; physical education; sport science; motion analysis; systematic review

Abstract

The rapid advancement of artificial intelligence (AI), particularly deep learning algorithms, has created significant opportunities for innovation in physical education and sport science. Deep learning enables high-precision analysis of human movement, biomechanical patterns, image recognition, and video-based performance assessment, thereby supporting data-driven instruction, motor skill evaluation, and adaptive learning environments. This study systematically reviews the implementation of deep learning technologies in physical education research and practice. The review employed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework to identify, screen, and analyze relevant studies published between 2015 and 2025. Data were collected from Scopus, Web of Science, and Google Scholar databases using keywords related to deep learning, physical education, motor learning, and sport analytics. The findings reveal that deep learning has been extensively applied in motion recognition, posture correction, skill classification, injury prevention analysis, and personalized feedback systems. Several studies also reported improvements in learning efficiency, student engagement, and accuracy of motor performance assessment through computer vision and wearable sensor integration [1], [2]. Nevertheless, the existing literature remains heavily concentrated on elite sports performance and higher education settings, while empirical applications in elementary and secondary school physical education are comparatively limited. In addition, challenges related to technological infrastructure, teacher digital competence, data privacy, and implementation costs continue to hinder broader adoption in school contexts [3], [4]. This review concludes that deep learning possesses substantial potential to transform physical education into a more adaptive, evidence-based, and technology-enhanced discipline. However, future studies should prioritize school-based interventions, longitudinal validation, and pedagogical integration models suitable for diverse educational environments. The novelty of this review lies in its emphasis on the educational implications of deep learning beyond competitive sport analytics, particularly within the context of inclusive and student-centered physical education.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Rahma Dewi
AU  - Imran Akhmad
AU  - Nurkadri Nurkadri
AU  - Bessy Sitorus Pane
AU  - Muhammad Reza Destya
PY  - 2026
DA  - 2026/06/24
TI  - Deep Learning Applications in Physical Education: A Systematic Review of Educational and Sport Science Perspectives
BT  - Proceedings of the 2nd International Conference of Sport Science, Sport Coaching Science, and Physical Education, and Recreation 2025 (ICOSSCOPER 2025)
PB  - Atlantis Press
SP  - 310
EP  - 323
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-591-1_30
DO  - 10.2991/978-2-38476-591-1_30
ID  - Dewi2026
ER  -